Difference between revisions of "Dhivehi and English/Structural transfer"

From LING073
Jump to: navigation, search
("Where is the bathroom?")
 
Line 72: Line 72:
  
 
  apertium-transfer: Rule 1 .<sent>/.<sent>
 
  apertium-transfer: Rule 1 .<sent>/.<sent>
^itg<itg>{^where<adv><itg>$}$ ^noun<SN><sg><def><dir>{^bathroom<n><sg>$}$^sent<SENT>{^?<sent>$}$^sent<SENT>{^.<sent>$}$</pre>
+
^itg<itg>{^where<adv><itg>$}$ ^noun<SN><sg><def><dir>{^bathroom<n><sg>$}$^sent<SENT>{^?<sent>$}$^sent<SENT>{^.<sent>$}$</pre>
  
 
<pre> $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng-interchunk
 
<pre> $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng-interchunk
 
   
 
   
 
  apertium-interchunk: Rule 2 itg<itg>{^where<adv><itg>$} noun<SN><sg><def><dir>{^bathroom<n><sg>$}
 
  apertium-interchunk: Rule 2 itg<itg>{^where<adv><itg>$} noun<SN><sg><def><dir>{^bathroom<n><sg>$}
^itg<itg>{^where<adv><itg>$}$ ^v<SV>{^be<vbser><pres><p3><sg>$}$ ^the<det>{^the<det><def><sp>$}$ ^noun<SN><sg><def><dir>{^bathroom<n><sg>$}$^sent<SENT>{^?<sent>$}$^sent<SENT>{^.<sent>$}$</pre>
+
^itg<itg>{^where<adv><itg>$}$ ^v<SV>{^be<vbser><pres><p3><sg>$}$ ^the<det>{^the<det><def><sp>$}$ ^noun<SN><sg><def><dir>{^bathroom<n><sg>$}$^sent<SENT>{^?<sent>$}$^sent<SENT>{^.<sent>$}$</pre>
  
 
<pre> $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng-postchunk
 
<pre> $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng-postchunk

Latest revision as of 09:39, 18 April 2019

Pre-evaluation

Statistics about input files
-------------------------------------------------------
Number of words in reference: 48
Number of words in test: 37
Number of unknown words (marked with a star) in test: 
Percentage of unknown words: 0.00 %

Results when removing unknown-word marks (stars)
-------------------------------------------------------
Edit distance: 47
Word error rate (WER): 97.92 %
Number of position-independent correct words: 2
Position-independent word error rate (PER): 95.83 %

Results when unknown-word marks (stars) are not removed
-------------------------------------------------------
Edit distance: 47
Word Error Rate (WER): 97.92 %
Number of position-independent correct words: 2
Position-independent word error rate (PER): 95.83 %

Statistics about the translation of unknown words
-------------------------------------------------------
Number of unknown words which were free rides: 0
Percentage of unknown words that were free rides: 0%

Definite Articles for Nouns

 $ echo ތަރައްގީ | apertium -d . div-eng
 the improvement
 $ echo ތަރައްގީ | apertium -d . div-eng-disam
 ^ތަރައްގީ<n><nhum><sg><def><dir>$^.<sent>$
 $ echo ތަރައްގީ | apertium -d . div-eng-biltrans 
 ^ތަރައްގީ<n><nhum><sg><def><dir>/improvement<n><sg><def><dir>$^.<sent>/.<sent>$
 $ echo ތަރައްގީ | apertium -d . div-eng-chunker
 
 apertium-transfer: Rule 3 ތަރައްގީ<n><nhum><sg><def><dir>/improvement<n><sg><def><dir>

 apertium-transfer: Rule 1 .<sent>/.<sent>
 ^noun<SN><sg><def><dir>{^improvement<n><sg>$}$^sent<SENT>{^.<sent>$}$
 $ echo ތަރައްގީ | apertium -d . div-eng-interchunk
 apertium-interchunk: Rule 1 noun<SN><sg><def><dir>{^improvement<n><sg>$}
 ^the<det>{^the<det><def><sp>$}$ ^noun<SN><sg><def><dir>{^improvement<n><sg>$}$^sent<SENT>{^.<sent>$}$
 $ echo ތަރައްގީ | apertium -d . div-eng-postchunk
 ^the<det><def><sp>$ ^improvement<n><sg>$^.<sent>$

"Where is the bathroom?"

 $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng
 where is the bathroom?
 $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng-disam
 ^ކޮބާ<itg>$ ^ފާހަނަ<n><nhum><sg><def><dir>$^؟<sent>$^.<sent>$
 $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng-biltrans
 ^ކޮބާ<itg>/where<adv><itg>$ ^ފާހަނަ<n><nhum><sg><def><dir>/bathroom<n><sg><def><dir>$^؟<sent>/?<sent>$^.<sent>/.<sent>$
 $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng-chunker
 
 apertium-transfer: Rule 2 ކޮބާ<itg>/where<adv><itg>

 apertium-transfer: Rule 3 ފާހަނަ<n><nhum><sg><def><dir>/bathroom<n><sg><def><dir>

 apertium-transfer: Rule 1 ؟<sent>/?<sent>

 apertium-transfer: Rule 1 .<sent>/.<sent>
 ^itg<itg>{^where<adv><itg>$}$ ^noun<SN><sg><def><dir>{^bathroom<n><sg>$}$^sent<SENT>{^?<sent>$}$^sent<SENT>{^.<sent>$}$
 $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng-interchunk
 
 apertium-interchunk: Rule 2 itg<itg>{^where<adv><itg>$} noun<SN><sg><def><dir>{^bathroom<n><sg>$}
 ^itg<itg>{^where<adv><itg>$}$ ^v<SV>{^be<vbser><pres><p3><sg>$}$ ^the<det>{^the<det><def><sp>$}$ ^noun<SN><sg><def><dir>{^bathroom<n><sg>$}$^sent<SENT>{^?<sent>$}$^sent<SENT>{^.<sent>$}$
 $ echo ކޮބާ ފާހަނަ؟ | apertium -d . div-eng-postchunk
 ^where<adv><itg>$ ^be<vbser><pres><p3><sg>$ ^the<det><def><sp>$ ^bathroom<n><sg>$^?<sent>$^.<sent>$

Post-Evaluation

Statistics about input files
-------------------------------------------------------
Number of words in reference: 48
Number of words in test: 49
Number of unknown words (marked with a star) in test: 
Percentage of unknown words: 0.00 %

Results when removing unknown-word marks (stars)
-------------------------------------------------------
Edit distance: 39
Word error rate (WER): 81.25 %
Number of position-independent correct words: 16
Position-independent word error rate (PER): 68.75 %

Results when unknown-word marks (stars) are not removed
-------------------------------------------------------
Edit distance: 39
Word Error Rate (WER): 81.25 %
Number of position-independent correct words: 16
Position-independent word error rate (PER): 68.75 %

Statistics about the translation of unknown words
-------------------------------------------------------
Number of unknown words which were free rides: 0
Percentage of unknown words that were free rides: 0%